Course · 7 chapters
Deep Learning
A decision-framework deep learning course for engineers. Choose PyTorch vs TensorFlow, judge depth vs classical ML, weigh transfer learning, and reason about CNNs. 7 chapters.
What you'll be able to do
- Choose between PyTorch and TensorFlow
- Know when deep learning beats classic ML
- Decide scratch training vs transfer learning
- Pick the right computer vision task
- Tune training hyperparameters
- Understand how CNNs see images
What's inside
- 1Frameworks de DL: PyTorch e TensorFlow
Um guia de decisão para os dois frameworks dominantes de deep learning — suas filosofias, ecossistemas e os fatores práticos que devem orientar sua escolha.
- 2Treinar do Zero vs Transfer Learning
A primeira decisão estratégica depois de escolher deep learning: treinar uma rede nova do zero ou aproveitar os ombros de gigantes pré-treinados.
- 3DL vs ML: Quando a Profundidade Vence
Um framework de decisão para escolher entre Deep Learning e Machine Learning clássico — com base nos seus dados, computação, prazos e necessidades de interpretabilidade.
- 4Deep Learning: Comece Aqui
Uma orientação de 12 minutos sobre o skill path de Deep Learning — por que ele existe, o que você vai construir, como os seis capítulos se conectam e por onde começar.
- 5Tipos de Tarefas de Visão
Classificação de imagens, detecção de objetos e segmentação — escolha a tarefa certa de visão computacional antes de escolher uma arquitetura.
- 6Técnicas de Treinamento para Deep Learning
O kit completo para treinar redes neurais — batch size, learning rates, funções de perda, ativações, otimizadores, regularização e early stopping.
- 7CNNs (Redes Neurais Convolucionais)
Projete, treine e interprete a arquitetura por trás da visão computacional moderna — da primeira convolução ao deploy em produção.
Frequently asked questions
- What will I learn in this deep learning course?
- You learn how to make the key decisions behind deep learning projects: choosing between PyTorch and TensorFlow, judging when depth beats classical ML, weighing transfer learning against training from scratch, selecting the right computer vision task, reasoning about training choices like learning rates and regularization, and understanding how CNNs are structured for vision.
- Is this a hands-on coding course or a conceptual one?
- It is a decision-framework course. The chapters focus on the trade-offs and reasoning behind framework choice, model strategy, vision tasks, and training, rather than walking through step-by-step coding labs. It assumes you can already write code on your own.
- Who is this deep learning path for?
- It is built for engineers and ML practitioners who want a clear mental model for deep learning decisions. It is part of the AI for Engineers track and assumes you are comfortable writing code and familiar with machine learning basics.
- How long is the course and is there a certificate?
- The path runs about 2.3 hours across 7 chapters, beginning with a short Start Here orientation chapter. When you finish the path you earn a certificate of completion you can share.
- Is this course free?
- No, this is a paid path available with an AI Academy by Anthropos subscription.
Earn a certificate
Complete all chapters to receive your certificate of completion.